In the present paper, a novel method was developed to predict the dynamic recrystallization (DRX) behavior during hot rolling based on machine learning in combination with physical models of niobium (Nb) micro-alloyed steels. It has been demonstrated that the comprehensive modeling could accurately predict the DRX behavior under different compositional and processing conditions, which not only exhibited higher precisions but also was in better agreement with physical metallurgical principles than traditional methods.

中文翻译：

---

Nb微合金钢的机器学习热变形行为及其对动态再结晶动力学的外推

在本文中，基于机器学习结合铌 (Nb) 微合金钢的物理模型，开发了一种预测热轧过程中动态再结晶 (DRX) 行为的新方法。结果表明，综合模型能够准确预测不同成分和加工条件下的DRX性能，与传统方法相比，不仅具有较高的精度，而且与物理冶金原理的一致性更好。